Which data sources are typically used in predictive maintenance to identify impending failures?

• A. Calendar-based scheduling and routine inspection.
• B. Vibration analysis, oil analysis, and thermal imaging.
• C. Visual inspection alone.
• D. Only operator observations.

Answer: (B)

Predictive maintenance relies on condition-monitoring data that show how a machine is actually behaving and when a fault is likely to occur. Vibration analysis captures changes in vibration patterns that reveal issues like bearing wear, imbalance, or misalignment. Oil analysis looks at wear metals, contaminants, and lubricant condition to indicate component wear and lubrication problems. Thermal imaging detects hot spots and uneven heating, pointing to overloads, poor lubrication, electrical resistance, or cooling failures. Together, these data sources provide objective, quantitative trends that can forecast failures before they happen, allowing maintenance to be scheduled just in time.

Relying on calendar-based scheduling uses fixed intervals and doesn’t reflect the machine’s real condition, so it may miss emerging problems or trigger unnecessary work. Visual inspection alone relies on human perception and may not detect internal or developing faults. Operator observations are subjective and can lag behind actual deterioration. The data-driven sources above give actionable insight into the machine’s health and are the foundation of predictive maintenance.